Gemini Omni, the ‘create anything’ model, starts today with lifelike video

Google announced Gemini Omni, a new multimodal generative model that can create realistic videos from text, images, video, and limited audio.

Gemini Omni Flash is available now for AI Plus subscribers and higher, while free creation via YouTube Shorts and YouTube Create is planned later.

Google also teased Omni Pro and said generated videos will include SynthID watermarks for transparency.

Some audio and speech-editing features are being delayed for safety reasons as Google expands its consumer-facing AI video push.